Intelligence knowledge graph optimization method and system based on graph neural network

Abstract

The invention discloses an intelligence knowledge graph optimization method and system based on a graph neural network, and the method comprises the steps: carrying out the topology of an intelligence knowledge graph, generating an intelligence knowledge graph topological graph, enabling nodes in the intelligence knowledge graph topological graph to represent the data points of satellite data, and enabling edges to represent the relation between the nodes; a feature updating model is set, when data points of new satellite data exist, features of the data points of the new satellite data are extracted, and corresponding nodes are updated according to the features of the data points of the new satellite data through the feature updating model; and setting an updated data accuracy evaluation model, performing data accuracy evaluation on the updated nodes, and feeding back an evaluation result to a user.

Description

Technical Field

[0001] The present invention belongs to the technical field of knowledge graph update optimization, and more specifically, to an intelligence knowledge graph optimization method and system based on graph neural network. Background Art

[0002] Graph Neural Networks (GNNs) is a neural network model based on graph-structured data, which is used to process and analyze data with complex relationships and interconnectivity.

[0003] Graph neural networks include: Graph: It is composed of nodes and edges. Nodes represent entities, and edges represent the relationship between entities.

[0004] Node Features: Each node has a feature vector that represents the properties or status of the node.

[0005] Edge Features: Each edge can also have a feature vector that represents the attributes or status of the edge.

[0006] However, there is no technical solution in the existing technology that can optimize the intelligence knowledge graph based on graph neural network. Summary of the invention

[0007] In order to solve the above technical problems, the present invention proposes an intelligence knowledge graph optimization method based on graph neural network, wherein the intelligence knowledge graph is a knowledge graph of satellite data, including: Topologically perform topology on the intelligence knowledge graph to generate an intelligence knowledge graph topology graph, wherein nodes in the intelligence knowledge graph topology graph represent data points of satellite data, and edges represent relationships between nodes; Setting a feature update model, when there is a new satellite data point, extracting the features of the new satellite data point, updating the model through the features, and updating the corresponding node according to the features of the new satellite data point; Set up an updated data accuracy assessment model, perform data accuracy assessment on the updated nodes, and feed back the assessment results to the user.

[0008] Furthermore, the feature update model includes: , in, The topological diagram of the intelligence knowledge graph Layer The characteristics of the nodes, is a gated recurrent unit, The topological diagram of the intelligence knowledge graph Layer The characteristics of the nodes, For the The set of all neighbor nodes of a node, is the number of attention heads, The topological diagram of the intelligence knowledge graph From the layer Node to The attention weight of each node, The topological diagram of the intelligence knowledge graph Layer The characteristics of a node.

[0009] Further, the topology of the computational intelligence knowledge graph From the layer Node to The attention weight of each node include: , in, is a leaky linear rectifier function, For the The transpose of the weights of the attention heads, For transposition, For the The node feature transformation matrix of the attention head, is the feature connection operator, For the The edge feature transformation matrix of the attention head is For the Node to The weight of the edge of the node, For the In the set of all neighbor nodes of node Layer The characteristics of a node.

[0010] Furthermore, the updated data accuracy assessment model includes: , in, To optimize the performance evaluation value, is the edge set of the intelligence knowledge graph topology, For the The current characteristics of the node, For the The current characteristics of the node, is the first hyperparameter, is the node set of the intelligence knowledge graph topology. For the The initial features of the nodes, is the second hyperparameter, For the The time loss function of each node.

[0011] Further, The time loss function of each node include: , in, is the total number of time steps, For the nodes at the time step Features, For the nodes at time step Features, is the third hyperparameter, For the nodes at time step characteristics.

[0012] The present invention also proposes an intelligence knowledge graph optimization system based on a graph neural network, wherein the intelligence knowledge graph is a knowledge graph of satellite data, including: A topology module, used to perform topology on the intelligence knowledge graph to generate an intelligence knowledge graph topology map, wherein the nodes in the intelligence knowledge graph topology map represent data points of satellite data, and the edges represent the relationships between the nodes; An updating module is used to set a feature updating model, and when there is a new data point of satellite data, extract the features of the new data point of satellite data, update the model through the features, and update the corresponding nodes according to the features of the new data point of satellite data; The evaluation module is used to set the updated data accuracy evaluation model, perform data accuracy evaluation on the updated nodes, and feed back the evaluation results to the user.

[0013] Furthermore, the feature update model includes: , in, The topological diagram of the intelligence knowledge graph Layer The characteristics of the nodes, is a gated recurrent unit, The topological diagram of the intelligence knowledge graph Layer The characteristics of the nodes, For the The set of all neighbor nodes of a node, is the number of attention heads, The topological diagram of the intelligence knowledge graph From the layer Node to The attention weight of each node, The topological diagram of the intelligence knowledge graph Layer The characteristics of a node.

[0014] Further, the topology of the computational intelligence knowledge graph From the layer Node to The attention weight of each node include: , in, is a leaky linear rectifier function, For the The transpose of the weights of the attention heads, For transposition, For the The node feature transformation matrix of the attention head, is the feature connection operator, For the The edge feature transformation matrix of the attention head is For the Node to The weight of the edge of the node, For the In the set of all neighbor nodes of node Layer The characteristics of a node.

[0015] Furthermore, the updated data accuracy assessment model includes: , in, To optimize the performance evaluation value, is the edge set of the intelligence knowledge graph topology, For the The current characteristics of the node, For the The current characteristics of the node, is the first hyperparameter, is the node set of the intelligence knowledge graph topology. For the The initial features of the nodes, is the second hyperparameter, For the The time loss function of each node.

[0016] Further, The time loss function of each node include: , in, is the total number of time steps, For the nodes at time step Features, For the nodes at time step Features, is the third hyperparameter, For the nodes at time step characteristics.

[0017] In general, the above technical solution conceived by the present invention has the following beneficial effects compared with the prior art: The present invention topologically performs the intelligence knowledge graph and generates an intelligence knowledge graph topological map, wherein the nodes in the intelligence knowledge graph topological map represent the data points of the satellite data, and the edges represent the relationship between the nodes; a feature update model is set, and when there are new satellite data points, the features of the new satellite data data points are extracted, and the corresponding nodes are updated through the feature update model and according to the features of the new satellite data data points; an updated data accuracy evaluation model is set, and the data accuracy of the updated nodes is evaluated, and the evaluation results are fed back to the user. Through the above technical scheme, the present invention can update the intelligence knowledge picture according to the new satellite data and evaluate it, thereby solving the problem of low accuracy and low efficiency of data update, and completing the knowledge graph optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is a flow chart of the method of embodiment 1 of the present invention; Figure 2 It is a structural diagram of the system of embodiment 2 of the present invention. DETAILED DESCRIPTION

[0019] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0020] The method provided by the present invention can be implemented in the following terminal environment, and the terminal may include one or more of the following components: a processor, a storage medium, and a display screen. The storage medium stores at least one instruction, and the instruction is loaded and executed by the processor to implement the method described in the following embodiment.

[0021] The processor may include one or more processing cores. The processor uses various interfaces and lines to connect various parts in the entire terminal, and executes various functions of the terminal and processes data by running or executing instructions, programs, code sets or instruction sets stored in the storage medium, and calling data stored in the storage medium.

[0022] The storage medium may include a random access memory (RAM) or a read-only memory (ROM). The storage medium may be used to store instructions, programs, codes, code sets or instructions.

[0023] The display screen is used to display the user interface of each application.

[0024] In addition, those skilled in the art can understand that the structure of the above terminal does not constitute a limitation on the terminal, and the terminal may include more or fewer components, or combine certain components, or arrange the components differently. For example, the terminal also includes components such as a radio frequency circuit, an input unit, a sensor, an audio circuit, and a power supply, which will not be described in detail here.

[0025] Example 1 like Figure 1 As shown, an embodiment of the present invention provides an intelligence knowledge graph optimization method based on a graph neural network, wherein the intelligence knowledge graph is a knowledge graph of satellite data, including: Step 101, topologically perform topology on the intelligence knowledge graph to generate an intelligence knowledge graph topology graph, wherein the nodes in the intelligence knowledge graph topology graph represent data points of satellite data, and the edges represent the relationships between the nodes; Step 102, setting a feature update model, when there is a new satellite data point, extracting the features of the new satellite data point, using the feature update model, and updating the corresponding node according to the features of the new satellite data point; Specifically, the feature update model includes: , in, The topological diagram of the intelligence knowledge graph Layer The characteristics of the nodes, is a gated recurrent unit, The topological diagram of the intelligence knowledge graph Layer The characteristics of the nodes, For the The set of all neighbor nodes of a node, is the number of attention heads, The topological diagram of the intelligence knowledge graph From the layer Node to The attention weight of each node, The topological diagram of the intelligence knowledge graph Layer The characteristics of a node.

[0026] Specifically, the computational intelligence knowledge graph topology diagram From the layer Node to The attention weight of each node include: , in, is a leaky linear rectifier function, For the The transpose of the weights of the attention heads, For transposition, For the The node feature transformation matrix of the attention head, is the feature connection operator, For the The edge feature transformation matrix of the attention head is For the Node to The weight of the edge of the node, For the In the set of all neighbor nodes of node Layer The characteristics of a node.

[0027] Step 103, setting an updated data accuracy assessment model, performing data accuracy assessment on the updated nodes, and feeding back the assessment results to the user.

[0028] Specifically, the updated data accuracy assessment model includes: , in, To optimize the performance evaluation value, is the edge set of the intelligence knowledge graph topology, For the The current characteristics of the node, For the The current characteristics of the node, is the first hyperparameter, is the node set of the intelligence knowledge graph topology. For the The initial features of the nodes, is the second hyperparameter, For the The time loss function of each node.

[0029] Specifically, The time loss function of each node include: , in, is the total number of time steps, For the nodes at time step Features, For the nodes at time step Features, is the third hyperparameter, For the nodes at time step characteristics.

[0030] Example 2 like Figure 2 As shown, an embodiment of the present invention further provides an intelligence knowledge graph optimization system based on a graph neural network, wherein the intelligence knowledge graph is a knowledge graph of satellite data, including: A topology module, used to perform topology on the intelligence knowledge graph to generate an intelligence knowledge graph topology map, wherein the nodes in the intelligence knowledge graph topology map represent data points of satellite data, and the edges represent the relationships between the nodes; An updating module is used to set a feature updating model, and when there is a new data point of satellite data, extract the features of the new data point of satellite data, update the model through the features, and update the corresponding nodes according to the features of the new data point of satellite data; Specifically, the feature update model includes: , in, The topological diagram of the intelligence knowledge graph Layer The characteristics of the nodes, is a gated recurrent unit, The topological diagram of the intelligence knowledge graph Layer The characteristics of the nodes, For the The set of all neighbor nodes of a node, is the number of attention heads, The topological diagram of the intelligence knowledge graph From the layer Node to The attention weight of each node, The topological diagram of the intelligence knowledge graph Layer The characteristics of a node.

[0031] Specifically, the computational intelligence knowledge graph topology diagram From the layer Node to The attention weight of each node include: , in, is a leaky linear rectifier function, For the The transpose of the weights of the attention heads, For transposition, For the The node feature transformation matrix of the attention head, is the feature connection operator, For the The edge feature transformation matrix of the attention head is For the Node to The weight of the edge of the node, For the In the set of all neighbor nodes of node Layer The characteristics of a node.

[0032] The evaluation module is used to set the updated data accuracy evaluation model, perform data accuracy evaluation on the updated nodes, and feed back the evaluation results to the user.

[0033] Specifically, the updated data accuracy assessment model includes: , in, To optimize the performance evaluation value, is the edge set of the intelligence knowledge graph topology, For the The current characteristics of the node, For the The current characteristics of the node, is the first hyperparameter, is the node set of the intelligence knowledge graph topology. For the The initial features of the nodes, is the second hyperparameter, For the The time loss function of each node.

[0034] Specifically, The time loss function of each node include: , in, is the total number of time steps, For the nodes at the time step Features, For the nodes at the time step Features, is the third hyperparameter, For the nodes at the time step characteristics.

[0035] Example 3 An embodiment of the present invention also proposes a storage medium storing multiple instructions, wherein the instructions are used to implement the intelligence knowledge graph optimization method based on graph neural network.

[0036] Optionally, in this embodiment, the above storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0037] Optionally, in this embodiment, the storage medium is configured to store program codes for executing the following steps: Step 101, topologically performing topology on the intelligence knowledge graph to generate an intelligence knowledge graph topology graph, wherein nodes in the intelligence knowledge graph topology graph represent data points of satellite data, and edges represent relationships between nodes; Step 102, setting a feature update model, when there is a new satellite data point, extracting the features of the new satellite data point, using the feature update model, and updating the corresponding node according to the features of the new satellite data point; Specifically, the feature update model includes: , in, The topological diagram of the intelligence knowledge graph Layer The characteristics of the nodes, is a gated recurrent unit, The topological diagram of the intelligence knowledge graph Layer The characteristics of the nodes, For the The set of all neighbor nodes of a node, is the number of attention heads, The topological diagram of the intelligence knowledge graph From the layer Node to The attention weight of each node, The topological diagram of the intelligence knowledge graph Layer The characteristics of a node.

[0038] Specifically, the computational intelligence knowledge graph topology diagram From the layer Node to The attention weight of each node include: , in, is a leaky linear rectifier function, For the The transpose of the weights of the attention heads, For transposition, For the The node feature transformation matrix of the attention head, is the feature connection operator, For the The edge feature transformation matrix of the attention head is For the Node to The weight of the edge of the node, For the In the set of all neighbor nodes of node Layer The characteristics of a node.

[0039] Step 103, setting an updated data accuracy assessment model, performing data accuracy assessment on the updated nodes, and feeding back the assessment results to the user.

[0040] Specifically, the updated data accuracy assessment model includes: , in, To optimize the performance evaluation value, is the edge set of the intelligence knowledge graph topology, For the The current characteristics of the node, For the The current characteristics of the node, is the first hyperparameter, is the node set of the intelligence knowledge graph topology. For the The initial features of the nodes, is the second hyperparameter, For the The time loss function of each node.

[0041] Specifically, The time loss function of each node include: , in, is the total number of time steps, For the nodes at time step Features, For the nodes at time step Features, is the third hyperparameter, For the nodes at time step characteristics.

[0042] Example 4 An embodiment of the present invention also proposes an electronic device, comprising a processor and a storage medium connected to the processor, wherein the storage medium stores a plurality of instructions, and the instructions can be loaded and executed by the processor so that the processor can execute the intelligence knowledge graph optimization method based on a graph neural network.

[0043] Specifically, the electronic device of this embodiment may be a computer terminal, and the computer terminal may include: one or more processors, and a storage medium.

[0044] Among them, the storage medium can be used to store software programs and modules, such as an intelligence knowledge graph optimization method based on a graph neural network in an embodiment of the present invention, and the corresponding program instructions / modules. The processor executes various functional applications and data processing by running the software programs and modules stored in the storage medium, that is, realizing the above-mentioned intelligence knowledge graph optimization method based on a graph neural network. The storage medium may include a high-speed random storage medium, and may also include a non-volatile storage medium, such as one or more magnetic storage systems, flash memory, or other non-volatile solid-state storage media. In some instances, the storage medium may further include a storage medium remotely arranged relative to the processor, and these remote storage media can be connected to the terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0045] The processor can call the information and application stored in the storage medium through the transmission system to perform the following steps: Step 101, topologically perform the intelligence knowledge graph to generate an intelligence knowledge graph topological graph, wherein the nodes in the intelligence knowledge graph topological graph represent data points of satellite data, and the edges represent the relationship between the nodes; Step 102, setting a feature update model, when there is a new satellite data point, extracting the features of the new satellite data point, using the feature update model, and updating the corresponding node according to the features of the new satellite data point; Specifically, the feature update model includes: , in, The topological diagram of the intelligence knowledge graph Layer The characteristics of the nodes, is a gated recurrent unit, The topological diagram of the intelligence knowledge graph Layer The characteristics of the nodes, For the The set of all neighbor nodes of a node, is the number of attention heads, The topological diagram of the intelligence knowledge graph From the layer Node to The attention weight of each node, The topological diagram of the intelligence knowledge graph Layer The characteristics of a node.

[0046] Specifically, the computational intelligence knowledge graph topology diagram From the layer Node to The attention weight of each node include: , in, is a leaky linear rectifier function, For the The transpose of the weights of the attention heads, For transposition, For the The node feature transformation matrix of the attention head, is the feature connection operator, For the The edge feature transformation matrix of the attention head is For the Node to The weight of the edge of the node, For the In the set of all neighbor nodes of node Layer The characteristics of a node.

[0047] Step 103, setting an updated data accuracy assessment model, performing data accuracy assessment on the updated nodes, and feeding back the assessment results to the user.

[0048] Specifically, the updated data accuracy assessment model includes: , in, To optimize the performance evaluation value, is the edge set of the intelligence knowledge graph topology, For the The current characteristics of the node, For the The current characteristics of the node, is the first hyperparameter, is the node set of the intelligence knowledge graph topology. For the The initial features of the nodes, is the second hyperparameter, For the The time loss function of each node.

[0049] Specifically, The time loss function of each node include: , in, is the total number of time steps, For the nodes at the time step Features, For the nodes at the time step Features, is the third hyperparameter, For the nodes at the time step characteristics.

[0050] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0051] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0052] In the several embodiments provided by the present invention, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the system embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0053] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0054] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0055] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only storage medium (ROM, Read-Only Memory), random access storage medium (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0056] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

Claims

1. A method for optimizing an intelligence knowledge graph based on a graph neural network, wherein the intelligence knowledge graph is a knowledge graph of satellite data, characterized in that: include: Topologically perform topology on the intelligence knowledge graph to generate an intelligence knowledge graph topology graph, wherein nodes in the intelligence knowledge graph topology graph represent data points of satellite data, and edges represent relationships between nodes; Setting a feature update model, when there is a new satellite data point, extracting the features of the new satellite data point, updating the model through the features, and updating the corresponding node according to the features of the new satellite data point; Set up an updated data accuracy assessment model, perform data accuracy assessment on the updated nodes, and feed back the assessment results to the user.

2. The intelligence knowledge graph optimization method based on graph neural network according to claim 1, characterized in that: The feature update model includes: ， in, The topological diagram of the intelligence knowledge graph Layer The characteristics of the nodes, is a gated recurrent unit, The topological diagram of the intelligence knowledge graph Layer The characteristics of the nodes, For the The set of all neighbor nodes of a node, is the number of attention heads, The topological diagram of the intelligence knowledge graph From the layer Node to The attention weight of each node, The topological diagram of the intelligence knowledge graph Layer The characteristics of a node.

3. The intelligence knowledge graph optimization method based on graph neural network as claimed in claim 2 is characterized in that: Computational Intelligence Knowledge Graph Topology From the layer Node to The attention weight of each node include: ， in, is a leaky linear rectifier function, For the The transpose of the weights of the attention heads, For transposition, For the The node feature transformation matrix of the attention head, is the feature connection operator, For the The edge feature transformation matrix of the attention head, For the Node to The weight of the edge between nodes, For the In the set of all neighbor nodes of node Layer The characteristics of a node.

4. The intelligence knowledge graph optimization method based on graph neural network as claimed in claim 3 is characterized in that: The updated data accuracy assessment model includes: ， in, To optimize the performance evaluation value, is the edge set of the intelligence knowledge graph topology, For the The current characteristics of the node, For the The current characteristics of the node, is the first hyperparameter, is the node set of the intelligence knowledge graph topology. For the The initial features of the nodes, is the second hyperparameter, For the The time loss function of each node.

5. The intelligence knowledge graph optimization method based on graph neural network as claimed in claim 4 is characterized in that: No. The time loss function of each node include: ， in, is the total number of time steps, For the nodes at time step Features, For the nodes at time step Features, is the third hyperparameter, For the nodes at time step characteristics.

6. An intelligence knowledge graph optimization system based on graph neural network, wherein the intelligence knowledge graph is a knowledge graph of satellite data, characterized in that: include: A topology module, used to perform topology on the intelligence knowledge graph to generate an intelligence knowledge graph topology map, wherein the nodes in the intelligence knowledge graph topology map represent data points of satellite data, and the edges represent the relationships between the nodes; An updating module is used to set a feature updating model, and when there is a new data point of satellite data, extract the features of the new data point of satellite data, update the model through the features, and update the corresponding nodes according to the features of the new data point of satellite data; The evaluation module is used to set the updated data accuracy evaluation model, perform data accuracy evaluation on the updated nodes, and feed back the evaluation results to the user.

7. The intelligence knowledge graph optimization system based on graph neural network according to claim 6, characterized in that: The feature update model includes: ， in, The topological diagram of the intelligence knowledge graph Layer The characteristics of the nodes, is a gated recurrent unit, The topological diagram of the intelligence knowledge graph Layer The characteristics of the nodes, For the The set of all neighbor nodes of a node, is the number of attention heads, The topological diagram of the intelligence knowledge graph From the layer Node to The attention weight of each node, The topological diagram of the intelligence knowledge graph Layer The characteristics of a node.

8. The intelligence knowledge graph optimization system based on graph neural network as claimed in claim 7, characterized in that: Computational Intelligence Knowledge Graph Topology From the layer Node to The attention weight of each node include: ， in, is a leaky linear rectifier function, For the The transpose of the weights of the attention heads, For transposition, For the The node feature transformation matrix of the attention head, is the feature connection operator, For the The edge feature transformation matrix of the attention head, For the Node to The weight of the edge between nodes, For the In the set of all neighbor nodes of node Layer The characteristics of a node.

9. The intelligence knowledge graph optimization system based on graph neural network as claimed in claim 8, characterized in that: The updated data accuracy assessment model includes: ， in, To optimize the performance evaluation value, is the edge set of the intelligence knowledge graph topology, For the The current characteristics of the node, For the The current characteristics of the node, is the first hyperparameter, is the node set of the intelligence knowledge graph topology. For the The initial features of the nodes, is the second hyperparameter, For the The time loss function of each node.

10. The intelligence knowledge graph optimization system based on graph neural network according to claim 9, characterized in that: No. The time loss function of each node include: ， in, is the total number of time steps, For the nodes at time step Features, For the nodes at time step Features, is the third hyperparameter, For the nodes at time step characteristics.

Images